Uncoupling of otolith and somatic 

 growth in Pagrus auratus (Sparidae) 



Malcolm R Francis 

 Maryann W. Williams 

 Andrea C. Pryce 

 Susan Pollard 

 Stephen G. Scott 



Fisheries Research Centre. MAF Fisheries 

 PO. Box 297. Wellington. New Zealand 



Slow-growing fish tend to have 

 heavier, larger otoliths than fast- 

 growing fish of the same length, be- 

 cause otoliths continue to grow even 

 when somatic growth has slowed or 

 stopped (e.g., Templeman & Squires 

 1956, Mosegaard et al. 1988, 

 Reznick et al. 1989, Secor & Dean 



1989, Secor et al. 1989, Campana 



1990, Pawson 1990). This uncou- 

 pling has important implications 

 for the back-calculation of fish 

 lengths from check marks in the 

 otoliths. If the relationship between 

 an otolith dimension and fish length 

 varies with growth rate, the back- 

 calculated lengths may be biased 

 (Campana 1990). This bias may be 

 largely overcome by specifying a 

 "biological intercept" (such as oto- 

 lith and somatic size-at-hatching) 

 and incorporating time-varying 

 growth (as measured by daily in- 

 crement widths) into the back- 

 calculation equation (Campana 

 1990). 



Pagrus auratus (Bloch & 

 Schneider 1801) is a commercially- 

 important sparid fish that ranges 

 through most of the temperate to 

 subtropical Western Pacific Ocean 

 (Paulin 1990). It has been reported 

 previously under a variety of syn- 

 onyms, especially P. major (Japan), 

 Chrysophrys auratus (Australia and 

 New Zealand), and C. unicolor (Aus- 

 tralia) (Paulin 1990). The common 

 name for P. auratus in New Zealand 



and Australia is "snapper," though 

 it is not a true snapper (Lutjanidae). 

 Uncoupling of otolith and somatic 

 growth has been demonstrated in 

 reared larval and presettlement 

 juvenile P. auratus from Japan 

 (Secor et al. 1989). In this study, 

 we report uncoupling of otolith and 

 somatic growth in wild, post- 

 settlement, juvenile New Zealand 

 snapper. We also discuss the impli- 

 cations this has for back-calculation 

 of juvenile snapper lengths using 

 otolith daily increments. 



Methods 



Snapper were caught using a small 

 otter trawl net equipped with a 

 20 mm stretched-mesh codend. 

 Samples were collected near Kawau 

 Island, Hauraki Gulf, New Zealand 

 (36°25'S, 174°46'E), January 1987 

 to March 1989. Fish were chilled 

 on capture, and frozen within 24 h. 

 After thawing, snapper were mea- 

 sured to the nearest mm fork length 

 (FL). Trial measurements before 

 and after freezing and thawing 

 showed that shrinkage was mini- 

 mal (mean shrinkage=2.03%, 

 SD=1.09%, n=42), thus no length 

 corrections were made. 



In New Zealand, snapper have a 

 prolonged summer spawning season 

 from October to February (Scott & 

 Pankhurst 1992), and we follow 



Paul (1976) in taking the theoreti- 

 cal birthday as 1 January. Each 

 year-class was numbered after its 

 first full year; e.g., snapper spawned 

 during the 1986-87 austral summer 

 were assigned to the 1987 year- 

 class. Age-0+ fish were identified 

 from length-frequency modes; they 

 grow to about 80-140 mmFL at the 

 end of their first year (Paul 1976; 

 M.R Francis, unpubl. data). 



Sagittae were removed, and one 

 of each pair was weighed and 

 measured for maximum length 

 ( anterioposterior axis) and height 

 (dorso- ventral axis). For snapper 

 <200 mmFL, transverse sections 1 

 were prepared from a subsample of 

 sagittae, and sulcal width was mea- 

 sured as the distance between the 

 sulcal side of the metamorphic mark 

 (Francis et al. In press) and the 

 sagitta margin at the ventral edge 

 of the sulcus. This measurement 

 was used in preference to total 

 sagitta width because the antisulcal 

 face of sagittae varied considerably 

 in shape, making it a poor refer- 

 ence surface, and because most of 

 the increase in sagitta width oc- 

 curred on the sulcal surface. The 

 collective term "size variables" is 

 used here when referring to sagitta 

 weight, length, height, and sulcal 

 width. 



A series of analyses of covariance 

 (ANCOVA) were used to investigate 

 the effects of year-class ( 1987 and 

 1988) and seven sampling periods 

 (Table 1) on the relationship be- 

 tween the four size-variables and 

 FL in 0+ snapper. 



Data from snapper samples col- 

 lected in Periods 2 and 3 (Table 1) 

 were used to determine whether 

 sagitta size at any given FL depends 



'Terminology used to describe otolith 

 planes and ageing follows Wilson et al. 

 (19871. 



Manuscript accepted 28 October 1992. 

 Fishery Bulletin, U.S. 91:159-164 (1993) 



159 



